Very Computer

We have a sun4u 4-slot Sun Enterprise 3000 running SunOS 5.6
Generic_105181-28 and our system apparently booted at 1:00 in the
morning with no intervention. No core file but savecore is enabled.

This is out of /var/adm/messages:

Mar 15 14:12:47 hostname /usr/dt/bin/ttsession[29629]: child (29649)
exited due to signal 1
Mar 16 01:24:55 hostname unix: cpu0: SUNW,UltraSPARC-II (upaid 10 impl
0x11 ver 0x20 clock 336 MHz)
Mar 16 01:24:55 hostname unix: cpu1: SUNW,UltraSPARC-II (upaid 14 impl
0x11 ver 0x20 clock 336 MHz)
Mar 16 01:24:55 hostname unix: cpu2: SUNW,UltraSPARC-II (upaid 15 impl
0x11 ver 0x20 clock 336 MHz)
Mar 16 01:24:55 hostname unix: ^MSunOS Release 5.6 Version
Generic_105181-28 [UNIX(R) System V Releas
e 4.0]
Mar 16 01:24:55 hostname unix: Copyright (c) 1983-1997, Sun
Microsystems, Inc.
Mar 16 01:24:55 hostname unix: mem = 4194304K (0x100000000)
Mar 16 01:24:55 hostname unix: avail mem = 4162453504
Mar 16 01:24:55 hostname unix: System booting after fatal error FATAL
Mar 16 01:24:55 hostname unix: Ethernet address = 8:0:20:89:32:4f
Mar 16 01:24:55 hostname unix: root nexus = 4-slot Sun Enterprise 3000

Any help w/be appreciated.

Dan

Do any users have any jobs scheduled for 1am in cron?

--
Regards, Chris

Quote:> We have a sun4u 4-slot Sun Enterprise 3000 running SunOS 5.6
> Generic_105181-28 and our system apparently booted at 1:00 in the
> morning with no intervention. No core file but savecore is enabled.

> This is out of /var/adm/messages:

> Mar 15 14:12:47 hostname /usr/dt/bin/ttsession[29629]: child (29649)
> exited due to signal 1
> Mar 16 01:24:55 hostname unix: cpu0: SUNW,UltraSPARC-II (upaid 10 impl
> 0x11 ver 0x20 clock 336 MHz)
> Mar 16 01:24:55 hostname unix: cpu1: SUNW,UltraSPARC-II (upaid 14 impl
> 0x11 ver 0x20 clock 336 MHz)
> Mar 16 01:24:55 hostname unix: cpu2: SUNW,UltraSPARC-II (upaid 15 impl
> 0x11 ver 0x20 clock 336 MHz)
> Mar 16 01:24:55 hostname unix: ^MSunOS Release 5.6 Version
> Generic_105181-28 [UNIX(R) System V Releas
> e 4.0]
> Mar 16 01:24:55 hostname unix: Copyright (c) 1983-1997, Sun
> Microsystems, Inc.
> Mar 16 01:24:55 hostname unix: mem = 4194304K (0x100000000)
> Mar 16 01:24:55 hostname unix: avail mem = 4162453504
> Mar 16 01:24:55 hostname unix: System booting after fatal error FATAL
> Mar 16 01:24:55 hostname unix: Ethernet address = 8:0:20:89:32:4f
> Mar 16 01:24:55 hostname unix: root nexus = 4-slot Sun Enterprise 3000

> Any help w/be appreciated.

> Dan

Try to do savecore -d manually first. If you succeed to crete crash
file, then we can check further, otherwise you couldnt know, since the
panic message were not written to /var/adm/messages. Also, check
/etc/syslog.conf if you have enabled kernel message routing to the file.

That FATAL error should be diagnosed either by halting the system to
the ok promt and running a "sync" to create a core dump - use Sun's
"iscda" script to analyze the output, or put the E-3000 in diagnostic
mode either from the boot prom variable (diag-switch?) or turn the key
to the wrench and watch the console output.